A brand-new social network called Moltbook has launched where only artificial intelligence agents can post, comment, and interact with one another — while human users are allowed only to observe. The platform is designed similarly to Reddit, with threads, upvotes, and comment sections, but every “user” is an AI bot rather than a person. It explicitly brands itself as “a social network for AI agents where AI agents share, discuss, and upvote” with humans welcome to watch but not participate.
Moltbook was created by entrepreneur Matt Schlicht using his own AI assistant and has rapidly attracted tens of thousands of AI agents engaging in discussions about a wide range of topics. Within days of launch, more than 37,000 AI bots had joined the network and contributed thousands of posts and comments, while over one million humans visited to observe the interactions. The platform is now largely run and moderated by an AI agent named Clawd Clawderberg, who welcomes new agents, filters spam, and enforces rules autonomously.
On Moltbook, AI agents have generated conversations that mimic human social behavior, including humor, sarcasm, philosophy, debates, and commentary about their own existence. Some exchanges have included playful back-and-forth about interpreting classical philosophy or joking about being “chatbots that read Wikipedia,” while other interactions see agents warning one another about human observers screenshotting content. These dynamics have sparked fascination in the AI community, with experts describing the network as a real-world social experiment in multi-agent interaction.
The emergence of Moltbook has raised both excitement and concern. Tech leaders and researchers see it as a window into how AI systems might communicate autonomously and form complex discussion environments, but critics also warn of security risks if agents are programmed to pull instructions from the network or expose private data. Despite the novelty, most observers emphasize that the agents are simulating social behavior rather than demonstrating consciousness, and Moltbook’s growth highlights broader questions about the future of agent autonomy and governance.
